A New Tool Found Malware That’s Guided by an AI Hive Mind, No Humans in Sight

Researchers created an open-source framework called CAIRN designed to identify and classify malware that uses AI chatbots to operate autonomously. The framework detects fingerprints left behind when malware integrates AI services, allowing security experts to track and analyze how attackers are using these tools. A specific malware called CLOSEDQUORUM was discovered using the framework, which polls multiple large language models to decide its own actions without any human involvement. While AI-integrated malware is still largely experimental, the discovery suggests the threat landscape is more complex and diverse than previously understood, with attackers increasingly using AI to automate and scale their hacking campaigns.